DROUGHT MONITORING INDICES

ANNEX D TO THE DROUGHT MITIGATION AND RESPONSE PLAN

August 2018

Colorado Drought Mitigation and Response Plan D.i

Annex D

August 2018

Table of Contents

1 Introduction ....................................................................................................1

1.1 Overview of Strengths and Limitations of Drought Indices for Colorado Drought

Monitoring............................................................................................................................................. 5

1.1.1 Standardized Precipitation Index (SPI) ........................................................................ 5

1.1.2 Palmer Drought Index (PDSI)....................................................................................... 6

1.1.3 Surface Water Supply Index (SWSI)............................................................................ 8

1.1.4 Colorado Monthly Precipitation and Percent of Normal Maps .................................. 9

1.1.5 Colorado Snowpack Accumulation and Ablation ..................................................... 10

1.1.6 Crop Moisture Index (CMI) ........................................................................................ 11

1.1.7 Keetch-Byram Drought Index (KBDI) ....................................................................... 12

2 Drought Indices for Future Consideration .................................................12

2.1.1 Evaporative Demand Drought Index (EDDI) ............................................................ 13

2.1.2 Standardized Precipitation-Evapotranspiration Index (SPEI)................................... 15

2.1.3 VegDRI and QuickDRI ............................................................................................... 16

3 Drought Impact Reporting and Conditions Data .......................................17

3.1.1 United States Drought Monitor (USDM) ................................................................... 17

3.1.2 NDMC Drought Impact Reporter ............................................................................... 18

3.1.3 CoCoRaHS Condition Monitoring Reports ............................................................... 19

3.1.4 USDA National Agricultural Statistics Service ......................................................... 19

3.2 2012-2013 Case Study of Drought Monitoring for the Arkansas River .............................. 20

4 Recommendations ........................................................................................24

5 References .....................................................................................................25

Colorado Drought Mitigation and Response Plan D.1

Annex D

August 2018

1 INTRODUCTION

Colorado’s diverse needs for timely and reliable water resources results in a broad spectrum of

stakeholders who hold different definitions of what constitutes drought. The great spatial variation

in precipitation and temperature regimes across the state presents a unique challenge to state

drought monitoring and planning. For these reasons, Colorado must apply a robust system of

drought monitoring tools to ensure drought conditions are adequately tracked for water user groups

throughout the state. To help new users become familiar with the drought monitoring tools

currently available and provide existing users with an up-to-date summary, this annex is designed

to provide a comprehensive outline of the most pertinent drought indices and their intended

applications.

This annex replaces what was in Annex D in the 2013 and 2010 updates of the Colorado Drought

Mitigation and Response Plan (Drought Plan or Plan). The annex to the previous versions of the

Plan included an analysis of the Palmer Drought Severity Index (PDSI) and its applicability to

Colorado, and discussed the modernization of the Surface Water Supply Index (SWSI). This

update of the annex in 2018 provides a synopsis of the current state of the art in drought indices.

Included is a discussion of their applicability in monitoring drought and activation of Annex A of

the Colorado's Drought Mitigation and Response Plan. Strengths and limitations of drought

indices are discussed. Other indices or remote sensing methods are discussed that, in the future,

could play a more formal role in drought monitoring and Plan activation. In addition, the

monitoring and reporting of drought impacts is also discussed.

The United States Drought Monitor (USDM), SWSI, PDSI and the Standardized Precipitation

Index (SPI) make up the key drought tracking products defined in the Plan. Guideline index values

for the SWSI, PDSI, and SPI and drought classifications from the USDM have been developed

and documented within previous versions of the Plan and are actively used in the decision-making

processes for the activation and deactivation of Plan Phases and Impact Task Force (ITF) groups.

These index guidelines attempt to provide a comprehensive estimate of drought conditions

(magnitude, duration, areal extent), while drought impact reports provide an essential observation-

based verification component necessary for the local and regional activation, mitigation, and

response decisions. The drought monitoring structure defined within the Plan is centered around

the professional judgment of the WATF to evaluate current and projected drought conditions using

drought index data and drought impact reports (Figure 1.1). By employing this convergence of

evidence approach, the Plan allows decision makers to examine a wealth of data to make timely

and informed drought mitigation and response decisions.

Colorado Drought Mitigation and Response Plan D.2

Annex D

August 2018

Figure 1.1 Workflow summary of the drought monitoring and response progression

outlined in the Plan

While the USDM, SWSI, PDSI, and SPI indices represent the most widely applied drought

tracking index tools, there are numerous other drought indices that have provided added benefit to

the state’s ongoing drought monitoring practices as well as several newer indices that may soon

provide further enhancements to drought monitoring in Colorado. A literature review of recent

publications provided the framework for a brief overview of the indices commonly applied within

the state and regional drought monitoring community. The Colorado Climate Center (CCC) and

National Drought Mitigation Center (NDMC) also provided expert knowledge regarding index

development and application for drought monitoring in Colorado. Each index summary includes a

breakdown of documented applications as well as some of the most relevant strengths and

weaknesses of the indices in their current state. By providing this information in an organized and

detailed manner, future updates to the Plan may continue to evaluate the list of indices and focus

efforts on linking local drought impact response/mitigation to the most appropriate drought indices

and index values.

This annex also presents a synopsis of the 2012-2013 drought conditions in southeastern Colorado

through a series of timeline plots. This high-level case study evaluation is intended to help illustrate

the drought progression and decision-making processes performed via the Plan. This analysis also

provides a simplified proof of concept example of a post-event evaluation that can be generated

for future Plan updates to further refine indices and threshold values for improved localized

monitoring.

The drought index “thresholds” outlined in Table 1.1 below were defined in previous drought

Plans (Annex A in the 2013 Plan) and are actively incorporated as the set of guidelines for drought

response activation and mitigation used by the WATF. The table includes a summary of the data

inputs to each index, time frame coverage, update frequency, and intensity scale range for the four

main drought indices along with the index value range guidelines for the defined drought phases.

This table was developed as a supplement to Table 1 in the Colorado Drought Response Plan

Annex A (2013).

Colorado Drought Mitigation and Response Plan D.3

Annex D

August 2018

Table 1.1 Outline of the primary drought indices currently documented in the 2013 Colorado Drought Plan

Drought Index

Primary Inputs Calculated or Effective Time

Frames

Update Frequency

Intensity Scale Typical Range

Index Value Range Guidelines (2013 Plan) Link to Data Access/Vie

w Normal Phase 1 Phase 2 Phase 3

USDM 40-50+ indicators and indices; flexible for future input/indices to be implemented

1 month to 12 months

every week

D0-D4 D0 Abnormally Dry

D1 Moderate Drought

D2 Severe Drought

D3-D4 Extreme to Exceptional Drought

NDMC UNL

PDSI Precipitation, Temperature

9 months every month

-4.0 to +4.0 (can exceed these bounds)

> -1.0 -1.0 to -2.0 < -2.0 -2.0 to -3.9 WRCC WWDT

SWSI Streamflow, Reservoir Storage, Forecasted Runoff (snowfall and precipitation)

3-6 months to several years

every month

-4.2 to +4.2

(can exceed these bounds)

+2.0 to -1.9 -2.0 to -2.9 -3.0 to -3.9

-4.0 to -4.9 CO DWR

SPI Precipitation Selectable; 1 month to 48 months+

daily -3.0 to +3.0 (can exceed these bounds)

> -0.5

(six month)

-0.6 to -1.0 < -1.0

(six month)

< -1.0 to -1.99 (six month)

HPRCC

A list of some of the most prominent and comprehensive drought toolboxes/dashboards and drought impact reporting/tracking resources

are provided below. Note that this list is just a sample of the many resources available for drought planning and monitoring.

Colorado Drought Mitigation and Response Plan D.4

Annex D

August 2018

Drought Indices Toolboxes and Dashboards

• CO DWR CDSS SWSI Mapper

• ESRI Living Atlas Drought Tracker

• High Plains Regional Climate Center Map Generator

• NDMC U.S. Drought Risk Atlas

• NIDIS Drought Portal Data Search

• NIDIS Summary of Drought Outlook & Forecast Products

• NIDIS/CCC Intermountain West DEWS

• NRCS Interactive Map

• USDM Current Conditions Products

• WRCC WestWide Drought Tracker Dashboard

• WWA Intermountain West Climate Dashboard

• NRCS Colorado Snow Survey

• CBRFC Water Supply Forecasts

Drought Impacts Reporting and Tracking

• CoCoRaHS Condition Monitoring Reports

• NDMC Drought Impact Reporter

• USDA National Agricultural Statistics Service (Colorado)

The general classification of each drought indicator/index is typically a direct reflection of the best

application of each drought monitoring product. It is important for users to have a basic

understanding of the purpose and limitations of each drought index before evaluating the output.

As a basic starting point, the drought research community commonly groups indicators and indices

into the following classifications:

• Meteorology (e.g. Standardized Precipitation Index)

• Soil moisture (e.g. Soil Moisture Deficit Index)

• Hydrology (e.g. Surface Water Supply Index)

• Vegetation (e.g. VegDRI)

The emergence of the term “flash drought” has gained substantial momentum within the drought

research community as well as the drought monitoring community. Otkin et al. (2017) recently

proposed a formalized definition of a “flash drought” to focus on the rate of rapid drought

intensification rather than the duration of drought conditions. By formalizing a clearer definition,

stakeholders who are susceptible to flash drought conditions can potentially be alerted to the initial

warning signs and forecasts for rapid drought intensification.

The next section of this annex provides a brief overview of the typical application of each index

for drought monitoring in Colorado.

Colorado Drought Mitigation and Response Plan D.5

Annex D

August 2018

1.1 Overview of Strengths and Limitations of Drought Indices for

Colorado Drought Monitoring

The following subsections summarize the applicability of each drought monitor as well as a brief

summary of the strengths and weaknesses identified in the literature review of recent publications

and discussions with drought monitoring experts. The index summaries are intended to provide a

high-level overview of the index purpose and application within the context of drought monitoring

in Colorado. Effective drought monitoring in Colorado requires a robust set of tools. While each

of the indices and monitoring products described below have valuable applications, it’s also

important to note where there is overlap between how indices are generated. When applying a

convergence of evidence approach to drought monitoring is important to understand how

individual index products are similar (e.g. precipitation input) and how they differ. Note that the

USDM is also an important component of drought monitoring in Colorado, but since it is a mixture

of other indices and drought impact monitoring products, it is discussed in Section 4 Drought

Impact Reporting and Conditions Data.

1.1.1 Standardized Precipitation Index (SPI)

Developed at the CCC, this index is broadly used to track and quantify conditions related to

meteorological drought. SPI uses historical

precipitation records to calculate a probability of

precipitation accumulation at timescales ranging

from 1 month to 48 months or longer.

1.1.1.1 Colorado Application

The range of SPI timescales makes the SPI a

robust product for evaluating meteorological

drought onset, intensity, and duration for

agriculture, water resources, and other sectors.

The shorter-timeframe SPI data can provide an

early indication of drought emergence which can

be useful for implementing drought mitigation

measures in a timely manner. Drought experts in

Colorado typically use the 60- and 90-day SPI to

look at more consistent patterns emerging and to

make initial recommendations to the USDM while the 6-month SPI is frequently used to look at

longer term patterns. It is also important to note that negative SPIs may carry more weight when

monitoring drought during a wet season vs. a dry season.

Figure 1.2 SPI image obtained from:

https://hprcc.unl.edu/maps.php?map=ACI

SClimateMaps#

Colorado Drought Mitigation and Response Plan D.6

Annex D

August 2018

1.1.1.2 Strengths

• SPI values for a range of timescale can be applied for multiple applications relating to the type

of drought impacts in question (e.g. shorter timeframe for agricultural drought and longer

timeframe for hydrologic drought).

• SPI is one of the easier indices to calculate and uses only precipitation data making it an ideal

index for regions with sparse data coverage.

• SPI values can also be compared across widely varied climates.

1.1.1.3 Weaknesses

• There is no temperature component in SPI so it does not directly account for evaporative

demand, an important component of the overall water balance. Drought events with similar

SPI may have different impacts because of differences in evaporative demand.

• The calculation can be sensitive the quantity and quality of the underlying historical

precipitation data used to generate the probability distribution (30+ years of historical data is

recommended).

• Selecting the appropriate time scale of SPI (e.g. 3-month vs. 6-month) can be challenging for

users when attempting to evaluate a range of potential drought vulnerabilities and impacts.

1.1.2 Palmer Drought Index (PDSI)

This prominent index was developed in the 1960s and has been widely applied to drought

monitoring practices. PDSI is designed to

use a simple supply and demand water

balance approach with inputs of

precipitation, temperature, and soil water

capacity estimates. There are also two

modified PDSI products commonly used

within drought monitoring practices:

• Self-Calibrated PDSI: uses a

mathematical calibration of model

coefficients to represent local climate

and soil properties

• Palmer Z: a measure of the relative

wetness/dryness anomalies of a region

evaluated against the full record at each

location; often preferred for its

normalized approach and better spatial comparison attributes

Figure 1.3: PDSI image obtained from:

https://wrcc.dri.edu/wwdt/index.php?folder=pdsi

Colorado Drought Mitigation and Response Plan D.7

Annex D

August 2018

1.1.2.1 Colorado Modified Palmer Drought Severity Index

Using the same water balance approach (temperature and precipitation inputs) as the PDSI, the

CMPDSI was designed to evaluate drought conditions at a finer spatial scale across Colorado. The

CMPDSI expanded on the original PDSI by enhancing the spatial resolution of the index from the

original 5 sub-regions (climate divisions) to 26 sub-regions to better represent the range of

topography and climate conditions within Colorado (Doesken et al. 1983). CCC discontinued the

generation of the CMPDSI in 2016 in favor of using similar products operationally available (e.g.

Western Regional Climate Center’s West Wide Drought Tracker PRISM PDSI). The newer

products use gridded climate data to provide a much higher resolution depiction of PDSI

conditions.

1.1.2.2 Colorado Application

PDSI considers both evapotranspiration as well as the moisture deficit within the soil column

making it a useful tool for detecting and quantifying meteorological and agricultural drought

conditions. This index was primarily developed for identifying drought conditions impacting the

agricultural sector, but its use has been expanded to other sectors (e.g. water supply). Within

Colorado, PDSI is primarily used for the eastern plains as its performance in the mountains is

largely inadequate.

1.1.2.3 Strengths

• PDSI is commonly applied around the world and has been extensively evaluated under a range

of conditions and has been thoroughly documented within the drought community literature.

• The application of soil properties and water balance methodology allow for a comprehensive

analysis of drought conditions compared to many other indices.

• As noted by Doesken & Ryan (2010), the CMPDSI has shown promising results for prediction

of winter wheat harvest as well as streamflow across various areas of the state.

1.1.2.4 Weaknesses

• PDSI has a relatively long memory (about 9 months) and does not respond quickly to emerging

dry conditions compared to other indices and is not ideal for short term evaluations of drought

conditions.

• It does not have a flexible multi-timescale calculation feature.

• The time lag between snowfall and snowmelt is not directly accounted for, all precipitation is

assumed to be immediately available.

Colorado Drought Mitigation and Response Plan D.8

Annex D

August 2018

1.1.3 Surface Water Supply Index (SWSI)

This index attempts to quantify the

amount of surface water available

in streams and reservoirs across the

state of Colorado. SWSI is

calculated for the seven major

intrastate basins as well as the HUC

8 basins throughout Colorado

(“Revised SWSI”). NRCS

originally developed this index in

1981 and modernized the SWSI for

Colorado as part of the Drought

Plan update effort in 2010. This

update included an improved spatial

resolution calculation (using the

HUC-8 statewide basins). The

Colorado Department of Water Resources (DWR) hosts a database of current and historical SWSI

statistics (included non-exceedance probabilities) for HUC8 basins in Colorado.

The Index uses the following inputs based on the time of year:

• January-June: SWSI = Streamflow Forecast + Reservoir Storage

• July-September: SWSI = Reservoir Storage + Previous Month’s Streamflow

• October-December: SWSI = Reservoir Storage

1.1.3.1 Colorado Application

This index is especially applicable for monitoring hydrologic drought conditions and month-to-

month fluctuations for water supply planning. SWSI was specifically designed to quantify drought

severity where water availability is driven by winter snow accumulation and ensuing snowmelt.

During the 2018 update of the Drought Plan, a member of the Colorado Water Availability Task

Force noted that SWSI has become decreasingly referenced in recent years; this could be the result

of the SWSI acronym confusion (the SWSI acronym also represents the Statewide Water Supply

Initiative).

1.1.3.2 Strengths

• SWSI takes into account the primary water supply components including snow accumulation,

snowmelt/runoff, and reservoir contents. This calculation provides a comprehensive evaluation

of the overall water balance of a given basin while accounting for manmade storage impacts

(reservoirs).

Figure 1.4: SWSI image obtained from:

https://gis.colorado.gov/dnrviewer/Index.html?viewer=

dwrhydrobaseviewer

Colorado Drought Mitigation and Response Plan D.9

Annex D

August 2018

• Several Colorado water supply organizations have implemented localized SWSI calculations

into their monitoring system, making SWSI a relatively familiar index.

1.1.3.3 Weaknesses

• The index must be recalculated anytime there are underlying changes to the input data (e.g.

reservoir capacity changes, diversion/return flow modifications, etc.).

• SWSI user’s have noted some basin-to-basin irregularities likely influenced by differing

reservoir and water supply management practices (at the Colorado HUC-8 scale). This may

lead to confusion or added uncertainty when attempting to evaluate conditions or trends

between neighboring basins.

1.1.4 Colorado Monthly Precipitation and Percent of Normal Maps

These indicators involve a simple statistical formula to generate using precipitation data as the

only input. This index can be applied quickly to evaluate meteorological drought conditions by

comparing recent precipitation values to mean historical data.

1.1.4.1 Colorado Application

Total precipitation and percent of normal

precipitation data is primarily used for simple

evaluations of meteorological drought

conditions. These variables are relatively

simple to understand and evaluate compared

to other indices.

1.1.4.2 Strengths

• These statistics can be easily generated to

cover a range of time periods in order to

evaluate a range of drought conditions.

1.1.4.3 Weaknesses

• Percent of normal values can be difficult

to compare among different climate

regimes (differing precipitation

climatology).

• “Normal” precipitation can be misunderstood when there are sizeable differences between the

median and mean for a given period.

Figure 1.5: Percent of normal precipitation

image obtained from:

https://hprcc.unl.edu/maps.php?map=ACIS

ClimateMaps#

Colorado Drought Mitigation and Response Plan D.10

Annex D

August 2018

1.1.5 Colorado Snowpack Accumulation and Ablation

Mountain snowpack in Colorado is a critical component of the hydrologic cycle. Higher elevation

snowfall and subsequent melt is a key driver of the overall water supply for many of the water

demands for both in-state and out-of-state users.

Snow Telemetry (SNOTEL) instrumentation

stations provide a relatively dense network of

data points to monitor snowpack conditions. The

Natural Resources Conservation Service (NRCS)

and National Water and Climate Center provide

current and historical data access for real time

evaluation of snowpack conditions.

The National Weather Service's National

Operational Hydrologic Remote Sensing Center

(NOHRSC) SNOw Data Assimilation System

(SNODAS) is also routinely used by water

supply managers and drought monitoring

systems. SNODAS produces high-resolution

(1km) gridded snowpack estimates through a

modelling and data assimilation system. This

data product merges ground observations (e.g.

SNOTEL point measurements) with satellite/airborne measurements, and model estimates of snow

cover.

1.1.5.1 Colorado Application

Maps and timeseries plots of the snowpack can be generated for SNOTEL stations as well as basin-

scale estimates via online tools. This data provides water supply managers and drought planners

with valuable estimates of the magnitude and timing of the snowmelt.

1.1.5.2 Strengths

• Snowpack data can provide and early indication of water shortages and potential drought

conditions.

• Daily to hourly temporal resolution of observed and modeled data products.

• Percent of normal snowpack information is a relatively easy product to understand for drought

planners and the public.

1.1.5.3 Weaknesses

• SNOTEL observations are point samples of an often complex and spatially variable snowpack

which may lead to an over or under estimate of true basin-wide snowpack.

Figure 1.6: Snow water equivalent percent

of normal image obtained from:

https://www.wcc.nrcs.usda.gov/snow/snow

_map.html

Colorado Drought Mitigation and Response Plan D.11

Annex D

August 2018

• Snowpack data should also be interpreted with reservoir storage levels and capacity to better

understand the expected water availability for downstream users.

• Gridded products like SNODAS have a limited history, therefore it’s difficult to assess how

far from “normal” the snowpack is at a given time when there is not a long enough record to

calculate a “normal”.

1.1.6 Crop Moisture Index (CMI)

This index was developed as a

compliment to the PDSI in an attempt to

address some of the shortcomings of

PDSI. CMI is intended to be a quick

responding index that evaluates the soil

column moisture deficit using estimates

of available moisture and potential

evapotranspiration.

1.1.6.1 Colorado Application

Mainly used for agricultural purposes, the

CMI gives a short-term status of a purely

agricultural drought which can change

rapidly. It is applicable in measuring

drought on a week to week basis for warm

season crops.

1.1.6.2 Strengths

• CMI is especially suited to evaluating drought conditions in relation to agriculture impacts.

1.1.6.3 Weaknesses

• As CMI is sensitive to quickly developing drought conditions, it may also produce a false sense

of drought recovery if longer-term moisture deficits are important.

• CMI may not be applicable for cool season or shallow rooted crops.

• Current spatial resolution is limited for Colorado (five climate divisions)

Figure 1.7: CMI image obtained from:

http://www.cpc.ncep.noaa.gov/products/analysi

s_monitoring/regional_monitoring/cmi.gif

Colorado Drought Mitigation and Response Plan D.12

Annex D

August 2018

1.1.7 Keetch-Byram Drought Index (KBDI)

KBDI was developed by the United States

Fire Service (USFS) as a fire risk index for fire

control managers. The index calculates the

moisture deficiency in duff and the upper soil

layers which is designed to evaluate the

flammability of organic material on the

ground. This measure is especially useful

when examining wildfire conditions and

potential drought stress on vegetation and

crops.

1.1.7.1 Colorado Application

KBDI estimates the amount of precipitation

necessary to return the soil to full field

capacity, with values ranging from 0 to 800

units (corresponding to 0 to 8 inches of water

in the soil). Wildfire potential is divided into

four levels based on KBDI values (0 to 800): low, moderate, high, and extreme contribution to fire

intensity. These levels correspond to typical seasonal variations.

1.1.7.2 Strengths

• The KBDI calculation is relatively simple index relying on daily maximum temperature and

daily precipitation as inputs.

1.1.7.3 Weaknesses

• The calculation assumes a limit of available moisture based on regionalized climate estimates

which may not be sufficient for every location.

2 DROUGHT INDICES FOR FUTURE CONSIDERATION

The following sections discuss some of the newer indices that have been the focus of recent

drought monitoring research and development. While these indices have started to gain attention

among Colorado drought monitoring experts, these products are still being tested and evaluated in

regards to operational capabilities. For this reason, the following indices are not included in the

formal drought monitoring guidelines; however, these indices can still provide a valuable resource

while the vetting process continues. Future updates to the Plan should reevaluate these indices for

updated information.

Figure 1.8: KBDI image obtained from:

http://www.wfas.net/index.php/keetch-byram-

index-moisture--drought-49

Colorado Drought Mitigation and Response Plan D.13

Annex D

August 2018

2.1.1 Evaporative Demand

Drought Index (EDDI)

EDDI is an experimental drought

monitoring tool specifically designed to

capture the early warning signs of water

stress through the emergence and/or

persistence of anomalous evaporative

demand. The EDDI calculation relies on

North American Land Data Assimilation

System (NLDAS) for gridded climate input

data. The CCC routinely examines the

EDDI products as part of their drought

monitoring toolbox. While the application

of EDDI to operation drought monitoring

is in an early stage at this point in time,

confidence is growing in that EDDI can be

a very useful tool for future drought plans

and early warning activation.

2.1.1.1 Future Colorado Application

EDDI is designed to be applicable for all land-cover types, which is especially useful for

Colorado’s diverse needs. The daily output (5-day lag) from EDDI provides a much-needed tool

for monitoring “flash drought” conditions which can have substantial impacts to the agricultural

sector in Colorado. EDDI can also be used as a fire-weather monitoring tool throughout the state.

2.1.1.2 Strengths

• Quick response time drought tool has long been a need for drought monitoring

• Capable of detecting “flash drought” conditions earlier than most other indices

• Drought category display colors can sharply highlight regions of drought concern

• EDDI model data have been validated using surface observations in Colorado

• EDDI is generated on the same range of timeframes as SPI

2.1.1.3 Weaknesses

• Still needs additional expert review and additional testing to fully understand and document

strengths and limitations

• Colored display of drought categories can be over emphasized under some conditions and

require user insight to fully evaluate (e.g. winter vs growing season anomalies)

• EDDI does not include any water supply information

Figure 2.1: EDDI image obtained from:

https://www.esrl.noaa.gov/psd/eddi/

Colorado Drought Mitigation and Response Plan D.14

Annex D

August 2018

Figure 2.2 below plots the 1-month to 12-month EDDI values (standardized anomalies in reference

evapotranspiration) for the 2010-2015 period. The data were extracted for the southeastern

quadrant of Colorado (spatially averaged output). This plot shows the early onset of dry conditions

starting in late 2010 and persisting into 2014 (positive index values). For perspective, plots the

EDDI timeseries for the 1990-2015 period. Note the anomalously dry conditions (magnitude and

duration) during the 2011-2014 drought period. Note that unlike other drought indices, raw EDDI

values are positive in dry conditions, and negative in wet conditions.

Figure 2.2: EDDI timeseries for southeastern Colorado for the 2010-2015 period

(data obtained from: https://www.esrl.noaa.gov/psd/eddi/)

Colorado Drought Mitigation and Response Plan D.15

Annex D

August 2018

Figure 2.3: EDDI timeseries for southeastern Colorado for the 1990-2015 period

(data obtained from: https://www.esrl.noaa.gov/psd/eddi/)

2.1.2 Standardized Precipitation-Evapotranspiration Index (SPEI)

This index is an extension of the SPI and

incorporates temperature data to account

for potential evapotranspiration in a basic

water balance calculation. SPEI aims to

improve the representation of drought

duration and magnitude resulting from

trends in potential evapotranspiration

change.

2.1.2.1 Future Colorado

Application

This index can be applied in all instances

where SPI is currently used while

providing a more reliable estimate of

drought conditions resulting from air

temperature influences.

2.1.2.2 Strengths

• Expected to be a more reliable

estimate over the traditional SPI

product especially when accounting for warming climate scenarios

Figure 2.4: SPEI Global Drought Monitor image

obtained from: http://spei.csic.es/map/maps.html

Colorado Drought Mitigation and Response Plan D.16

Annex D

August 2018

• Available on the same range of timeframes as SPI (<1-month to > 48-months+)

2.1.2.3 Weaknesses

• Real-time SPEI output is currently based on the Thornthwaite equation for potential

evapotranspiration (temperature-based estimate) which can have limited reliability in western

US applications

• Output is highly sensitive to the method used to calculate the PET product (e.g. Thornthwaite

vs. Penman Monteith) especially in regions where wind can play a big role in varying

evapotranspiration

• Not yet extensively tested and evaluated by the drought monitoring community

2.1.3 VegDRI and QuickDRI

The Vegetation Drought Response

Index (VegDRI) and the Quick

Drought Response Index (QuickDRI)

and have been developed by the U.S.

Geological Survey (USGS), Earth

Resources Observation and Science

(EROS) Center and the National

Drought Mitigation Center (NDMC).

These hybrid modeled drought indices

incorporate satellite-based

observations of vegetation properties,

climate data, and other biophysical

data products to quantify the current

vegetative state. VegDRI has a

seasonal time horizon for

characterizing drought conditions,

while QuickDRI is designed for

detecting early onset and rapidly

developing drought conditions on a

roughly 1-month timescale (i.e. flash

drought conditions).

2.1.3.1 Future Colorado Application

These two indices provide a high-spatial resolution (1km) drought indicator specifically designed

for vegetation stress and agricultural applications. VegDRI has provided valuable updates on the

mid to late summer vegetation health within Colorado whereas the early growing season output is

largely just a function of the input SPI blend. (personal communication with CCC, Jan 2018).

Figure 2.5: VegDRI image obtained from:

https://vegdri.cr.usgs.gov/viewer/

Colorado Drought Mitigation and Response Plan D.17

Annex D

August 2018

2.1.3.2 Strengths

• Both indices are a weekly product

• 1km spatial resolution

• Both indices are a hybrid index – incorporating numerous variables representing the hydrologic

cycle and drought-related vegetation stress

2.1.3.3 Weaknesses

• Only applicable in areas with vegetation cover and during the growing season

• Relies on precipitation as an input which may lead to redundancies in output with other

precipitation driven indices (e.g. SPI & PDSI)

3 DROUGHT IMPACT REPORTING AND CONDITIONS DATA

Comprehensive drought impact assessments are a well-known shortcoming within the drought

monitoring community. Drought management decisions are ultimately focused on mitigating and

responding to a range of drought related impacts; however, localized impact assessments are often

complicated by a multitude of socioeconomic and physical factors relating to drought vulnerability

and exposure. Aligning projected drought conditions with known impacts remains a critical link

to developing a localized drought mitigation and response system.

While historical records of drought-related impacts have typically focused heavily on agricultural

impacts, several drought impact reporting systems and databases have emerged in recent years to

help meet the growing need for a comprehensive record of current and historical drought impacts

across a range of sectors. Three resources are described in the following sections in an effort to

continue to build usership and emphasize their application for continued improvements to drought

mitigation efforts in Colorado.

3.1.1 United States Drought

Monitor (USDM)

The USDM was the first operational

composite-based drought monitoring

product used in the US. While it is

primarily based on a handful of key

indicators (SPI at multiple time scales,

PDSI, modeled soil moisture, weekly

streamflow) it uses a flexible

framework to incorporate up to 50+

inputs from a variety of sources with

capabilities to continuously

incorporate newly developed

Figure 3.1: USDM image obtained from:

http://droughtmonitor.unl.edu/CurrentMap.aspx

Colorado Drought Mitigation and Response Plan D.18

Annex D

August 2018

data/indices as they become available. For example, snow-water equivalent is incorporated in the

wester US in the winter. The weekly map generation process also relies on human observations

and impact reports. The Colorado Climate Center continues to provide and coordinate drought

monitoring data and information input from Colorado water experts to the USDM on a weekly

basis.

3.1.1.1 Colorado Application

The USDM is often a key focal point of drought monitoring efforts covering a wide range of

applications throughout the state of Colorado. The USDM is not limited to seasonal interpretation

(e.g. wet season or growing season) and can be consistently evaluated throughout the year. The

WATF includes the percent coverage of drought categories at all monthly meetings.

3.1.1.2 Strengths

• USDM incorporates numerous indicators and indices covering a range of timescales to produce

a robust drought identification and intensity classification.

• The five-scale drought intensity classifications are widely used throughout the country and

familiar to stakeholders, the media, and the public. Other indices attempt to apply the same

general intensity scale for data displays.

3.1.1.3 Weaknesses

• Very localized drought conditions may not be sufficiently represented within the USDM

resolution

• Drought can only be depicted where the actual water shortage is present, not necessarily where

the area of impacts is greatest. For example, a

drought over the mountains (where the majority of

the water resources reside) can have major

implications over many other regions, including

other states, but the USDM does not provide that

information.

3.1.2 NDMC Drought Impact Reporter

The Drought Impact Reporter (DIR) was initially

released in 2005 as one of the first nationwide drought

impact databases. This data catalog incorporates

drought-related reports from a wide variety of sources

including media reports, user-submitted reports,

CoCoRaHS reports, NWS Drought Information

Statements, and state agency reports. DIR

documentation notes that media reports of drought-

related news stories make up the largest contribution to

Figure 3.2: Drought Impact Reporter

map interface

(http://droughtreporter.unl.edu/map/)

Colorado Drought Mitigation and Response Plan D.19

Annex D

August 2018

the database; however, user-submitted reports from the public are also an essential component of

the database. The DIR is also designed to group drought impacts by relevant characteristics such

as the drought category (e.g. agriculture, water supply/quality, tourism/recreation), drought

duration, and affected locations.

3.1.3 CoCoRaHS Condition Monitoring Reports

Community Collaborative Rain Hail and

Snow (CoCoRaHS) is a network of

community volunteers who provide

detailed measurements and observations

of daily weather conditions nationwide

(and international). Started in Fort

Collins in 1998, the network has grown

to include thousands of active daily

reports that provide a valuable database

of localized weather observations. While

initially designed to primarily record

precipitation observations, the

CoCoRaHS reporting system has adapted

a condition monitoring reporting system.

The condition monitoring reports submitted by

CoCoRaHS volunteers are designed to provide a

constant stream of qualitative reports regarding the

status of wet or dry conditions and the localized

impacts. Condition monitoring reports provide

date stamped observations with a relative

conditions scale bar (ranging from severely dry to

near normal to severely wet) as well as a

description of the impacts affecting the reporter.

The condition monitoring reports also include a

classification structure similar to the DIR to help

streamline the categorization of impacts by sectors

(e.g. agriculture, fire, business etc.). Processed

reports can be viewed through an interactive map

interface or by a database search query.

3.1.4 USDA National Agricultural

Statistics Service

The United States Department of Agriculture

(USDA) National Agricultural Statistics Service

(NASS) maintains a detailed record of crop,

Figure 3.3: CoCoRaHS Condition Monitoring

map interface

(https://www.cocorahs.org/Maps/conditionm

onitoring/#)

Figure 3.4: Secretarial county drought

designations for 2017

(https://www.usda.gov/sites/default/file

s/documents/usda-drought-fast-track-

designations.pdf)

Colorado Drought Mitigation and Response Plan D.20

Annex D

August 2018

livestock, pasture, and range conditions relating to drought impacts. With water availability

playing such a significant role in the agricultural sector, crop conditions are routinely documented

for large portions of the country through annual, monthly, and weekly reports. County and

statewide crop progress reports provide records regarding commodity conditions, soil moisture

conditions, and crop yields. The USDA also maintains a record of Secretarial Drought

Designations by county.

3.2 2012-2013 Case Study of Drought Monitoring for the Arkansas

River

The 2011-2013 drought conditions impacted the entire state of Colorado, but conditions were

especially severe in the southeast quadrant of the state which includes most of the Upper Arkansas

River basin. A simple case study re-analysis was developed for the 2011-2015 period to help

illustrate the drought progression and subsequent drought mitigation and response actions applied

throughout the state. Figure 3.5 is a snapshot of the USDM during the week of February 18, 2013

showing the widespread drought conditions over Colorado with exceptional drought conditions

over a large portion of the eastern part of the state.

Figure 3.5: ESRI Drought Tracker showing the USDM drought severity in early

2013.

This case study aims to provide a simple synopsis of the drought conditions by plotting the

historical drought indices timeseries along with a timeline of the documented drought

mitigation/response actions, and drought impact reports. Figure 3.6, Figure 3.7, and Figure 3.8

outline the evolution of drought conditions evident by the USDM, several drought index

timeseries, and a timeline of the drought Plan actions.

For simplicity, the drought indices datasets focus on the spatial average for the Upper Arkansas

River basin (HUC 110200). The USDM drought index percent coverage time series plot for the

Colorado Drought Mitigation and Response Plan D.21

Annex D

August 2018

Upper Arkansas basin was obtained from the USDM webpage. The historical timeseries of PDSI,

Palmer Z, SC-PDSI, SPI (1-month), and SPEI (1-month) were obtained from the WRCC

WestWide Drought Tracker (Abatzoglou et al., 2017). The WestWide Drought Tracker dataset

applies the monthly gridded PRISM data products as the foundation for the calculation of the

monthly historical index timeseries. Recall that the range of typical index values may differ from

one index to the next, and the user should give special attention to the magnitude of each index

when quantifying the overall drought signal. The SWSI timeseries data for the Arkansas basin

were obtained from the Colorado DWR Information Marketplace. Drought Plan

activation/deactivation dates and descriptions were mostly attained from monthly WATF Drought

Update Reports and the Governor’s Memorandum documents (all archived files are available on

CWCB Laserfiche WebLink database).

Figure 3.6: USDM Drought categories percent coverage for the Upper Arkansas

(Climate Division 501) for the 2011-2015 period (data obtained from

http://droughtmonitor.unl.edu)

Colorado Drought Mitigation and Response Plan D.22

Annex D

August 2018

Figure 3.7: Monthly time series plots of drought index values for six drought

indices during the 2011-2015 period (data obtained from the West Wide Drought

Tracker and the Colorado DNR)

Figure 3.8: Timeline of drought Plan actions/phases including ITF activation and

deactivation

By incorporating a variety of drought indices, a “convergence of evidence” approach can be

applied for sound drought decision making. The USDM and drought indices timeseries data plots

above attempt to illustrate the quantitative data available to the WATF and drought decision

makers as the drought progressed; however; qualitative data sources are also important for

identifying and validating drought impacts. Qualitative data includes information such as drought

impact observations from a variety of sectors. These observations are essential for focusing

drought mitigation and response resources prior-to and during a drought while also providing an

Colorado Drought Mitigation and Response Plan D.23

Annex D

August 2018

invaluable dataset to evaluate drought indices and refine future drought planning resources at a

localized scale.

Figure 3.9: Summary of the 2011-2015 drought impact reports for Colorado

Figure 3.9 summarizes the monthly count of new drought reports categorized into agricultural and

water supply/quality impacts. Impact data were obtained from the NDMC Drought Impact

Reporter for the entire state of Colorado.

Individual impact reports include a

description, start and end date, and

category classification tag(s). The data

search yielded 213 reports that have a

start date, end date, or timespan that

occurs within the 2011-2015 period. It’s

important to note that the figure above

represents the start date of each report and

drought impacts may have extended

months to years into the future. Numerous

reports (78) did not include an ending date

to fully examine the temporal component

of drought impacts, and this finding

further illustrates the continued need for

comprehensive condition monitoring

impact reporting prior to, during, and after

drought periods. For added spatial

reference, Figure 1.23 is a web map

screenshot of the NDMC Drought Impact

Figure 3.10: NDMC Drought Impact Reporter

summary of county impact counts for the 2012-

2014 period.

Colorado Drought Mitigation and Response Plan D.24

Annex D

August 2018

Reporter showing the total drought impact report counts by Colorado County for the 2012-2014

period.

4 RECOMMENDATIONS

After extensive discussions with leading experts in the drought monitoring field as well as

representatives from the Colorado Impact Task Forces a list of key recommendations for continued

development and improvement to the Colorado Drought Plan was developed.

1. Recommend the Water Availability Task Force and each individual Impact Task Force

expand efforts to document drought impacts throughout Colorado as well as any mitigation

actions or response measures put in place. Comprehensive records of dates, locations, and

drought-related impact descriptions are difficult to come by but are extremely valuable for

continued improvement to understanding drought vulnerabilities at a localized scale. A

robust drought impact database can be combined with archived drought index data to

provide a foundation for a future validation and refinement of drought monitoring practices

and responses. Results of future validation efforts can be used to focus future drought Plan

updates on linking drought vulnerabilities to specific drought indices (and index values) by

location and stakeholder groups. Drought Plan Appendix B Actions Taken to Reduce

Drought Impacts in Previous could be used as a starting point and should be updated

periodically during future droughts.

2. Continue to monitor/evaluate EDDI and other newer indices currently being tested for

future Plan implementation as they mature.

3. Consider future implementation of drought index guidelines as a function of both index

values and index percentile ranking to allow for a streamlined comparison among the

different indices.

4. Update and modernize the CWCB Drought Planning Toolbox website by providing newer

resources to data hosting portals and dashboards.

Colorado Drought Mitigation and Response Plan D.25

Annex D

August 2018

5 REFERENCES

Abatzoglou, J. T., McEvoy, D. J., & Redmond, K. T. (2017). The West Wide Drought Tracker:

Drought monitoring at Fine Spatial Scales. Bulletin of the American Meteorological Society,

98(9), 1815–1820. http://doi.org/10.1175/BAMS-D-16-0193.1

Alley, W. M. (1984). The Palmer Drought Severity Index: Limitations and Assumptions. Journal

of Climate and Applied Meteorology, 23(7), 1100–1109. http://doi.org/10.1175/1520-

0450(1984)023<1100:TPDSIL>2.0.CO;2

Bachmair, S., Stahl, K., Collins, K., Hannaford, J., Acreman, M., Svoboda, M., … Overton, I. C.

(2016). Drought indicators revisited: the need for a wider consideration of environment and

society. Wiley Interdisciplinary Reviews: Water, 3(4), 516–536. http://doi.org/10.1002/wat2.1154

Doesken, N. J., & Mckee, T. B. (1991). Development of a Surface Water Supply Index for the

Western United States; Climatology Report #91-3. Fort Collins, CO.

Doesken, N., & Ryan, W. (2010). The Development and Evolution of Best Practice Strategies for

Using Drought Monitoring Tools in Colorado; Prepared October 2010 for the Colorado Water

Conservation Board. Fort Collins, CO.

Finnessey, T., Hayes, M., Lukas, J., & Svoboda, M. (2016). Using climate information for drought

planning. Climate Research, 70(2–3), 251–263. http://doi.org/10.3354/cr01406

Fu, X., Svoboda, M., Tang, Z., Dai, Z., & Wu, J. (2013). An overview of US state drought plans:

Crisis or risk management? Natural Hazards, 69(3), 1607–1627. http://doi.org/10.1007/s11069-

013-0766-z

Hayes, D. M. J., Alvord, C., & Lowrey, J. (2002). Drought Indices. International Journal of

Climatology, 23 (July), 18. http://doi.org/10.1002/joc.931

Hobbins, M. T., Wood, A., McEvoy, D. J., Huntington, J. L., Morton, C., Anderson, M., & Hain,

C. (2016). The Evaporative Demand Drought Index. Part I: Linking Drought Evolution to

Variations in Evaporative Demand. Journal of Hydrometeorology, 17(6), 1745–1761.

http://doi.org/10.1175/JHM-D-15-0121.1

Littell, J. S., Peterson, D. L., Riley, K. L., Liu, Y., & Luc, C. H. (2016). A Review of the

Relationships Between Drought and Forest Fire in the United States. Global Change Biology, 1–

17. http://doi.org/10.1111/gcb.13275

McEvoy, D. J., Huntington, J. L., Hobbins, M. T., Wood, A., Morton, C., Anderson, M., & Hain,

C. (2016). The Evaporative Demand Drought Index. Part II: CONUS-Wide Assessment against

Common Drought Indicators. Journal of Hydrometeorology, 17(6), 1763–1779.

http://doi.org/10.1175/JHM-D-15-0122.1

Colorado Drought Mitigation and Response Plan D.26

Annex D

August 2018

Mckee, T. B., Doesken, N. J., Kleist, J., Shrier, C. J., & Stanton, W. P. (2000). A History of

Drought in Colorado: Lessons learned and what lies ahead. Water in Balance, (9).

NDMC. (2017). Fact Sheet: Quick Drought Response Index. Retrieved from

http://drought.unl.edu/archive/Documents/QuickDri/QuickDriFactSheet.pdf

Otkin, J. A., Anderson, M. C., Hain, C., Mladenova, I. E., Basara, J. B., & Svoboda, M. (2013).

Examining Rapid Onset Drought Development Using the Thermal Infrared–Based Evaporative

Stress Index. Journal of Hydrometeorology, 14(4), 1057–1074. http://doi.org/10.1175/JHM-D-12-

0144.1

Otkin, J. A., Anderson, M. C., Hain, C., Svoboda, M., Johnson, D., Mueller, R., … Brown, J.

(2016). Assessing the evolution of soil moisture and vegetation conditions during the 2012 United

States flash drought. Agricultural and Forest Meteorology, 218–219, 230–242.

http://doi.org/10.1016/j.agrformet.2015.12.065

Otkin, J. A., Svoboda, M., Hunt, E. D., Ford, T. W., Anderson, M. C., Hain, C., & Basara, J. B.

(2017). Flash Droughts: a Review and Assessment of the Challenges Imposed By Rapid Onset

Droughts in the United States. Bulletin of the American Meteorological Society, BAMS-D-17-

0149.1. http://doi.org/10.1175/BAMS-D-17-0149.1

Palmer, W. C. (1965). Meteorological Drought. U.S. Weather Bureau, Res. Pap. No. 45. Retrieved

from https://www.ncdc.noaa.gov/temp-and-precip/drought/docs/palmer.pdf

Rangwala, I., Barsugli, J., & Dewes, C. (2015). EDDI A Powerful Tool for Early Drought

Warning. Retrieved from http://wwa.colorado.edu/publications/reports/EDDI_2-pager.pdf

Sivakumar, V.K., M., Motha, R. P., Wilhite, D. A., & Wood, D. A. (2010). Quantification of

Agricultural Drought for Effective Drought Mitigation and Preparedness: Key Issues and

Challenges. In Proceedings of the WMO/UNISDR Expert Group Meeting on Agricultural Drought

Indices (p. 197). Murcia, Spain: World Meteorological Organization. Retrieved from

http://www.wamis.org/agm/pubs/agm11/agm11.pdf

Steinemann, A. C., Hayes, M. J., & Cavalcanti, L. F. N. (2005). Drought and Water Crises Science,

Technology, and Management Issues. (D. A. Wilhite, Ed.). Boca Raton, FL: CRC Press. Retrieved

from http://www.oregon.gov/owrd/docs/HB4113/AM_wilhite_buchanan_2005_drought hazard

understanding natrual and social context.pdf

Steinemann, A. C., Cavalcanti, L. F. N., Asce, M., & Cavalcanti, L. F. N. (2006). Developing

Multiple Indicators and Triggers for Drought Plans. Journal of Water Resources Planning and

Management, 132(3), 164–174. http://doi.org/10.1061/(ASCE)0733-9496(2006)132:3(164)

Colorado Drought Mitigation and Response Plan D.27

Annex D

August 2018

Svoboda, M. D., Fuchs, B. A., Poulsen, C. C., & Nothwehr, J. R. (2015). The drought risk atlas:

Enhancing decision support for drought risk management in the United States. Journal of

Hydrology, 526, 274–286. http://doi.org/10.1016/j.jhydrol.2015.01.006

Vicente-Serrano, S. M., Beguería, S., & López-Moreno, J. I. (2010). A multiscalar drought index

sensitive to global warming: The standardized precipitation evapotranspiration index. Journal of

Climate, 23(7), 1696–1718. http://doi.org/10.1175/2009JCLI2909.1

Wood, A. W. (2008). The University of Washington Surface Water Monitor: An experimental

platform for national hydrologic assessment and prediction. In American Meteorological Society

Annual Meeting (pp. 1–13). New Orleans. Retrieved from

https://ams.confex.com/ams/pdfpapers/134844.pdf?origin=publication_detail

World Meteorological Organization and Global Water Partnership, Svoboda, M., & Fuchs, B.

(2016). Handbook of drought indicators and indices.Integrated Drought Management Programme

(IDMP), Integrated Drought Management Tools and Guidelines Series 2.

http://doi.org/10.1007/s00704-016-1984-6

Zargar, A., Sadiq, R., Naser, B., & Khan, F. I. (2011). A review of drought indices. Environmental

Reviews, 19(NA), 333–349. http://doi.org/10.1139/a11-013